Research projects

The BioMetaSEP consortium proposes a prospective spatio-temporal analysis of metagenomic biomarkers of abundance and diversity of opportunistic pathogens in the intestinal and respiratory microbiota. The objective is to quantify the risk of sepsis (based on hypercolonization pressure), to predict the success of the antimicrobial therapy (based on the intestinal and/or respiratory resistomes), and to determine the influence of interventions (eg. use of antibiotics, enteral and parenteral nutrition, immunosuppressants, catheters and other devices, mechanical ventilation) in the dynamics of the individual microbiomes. The experimental approach relies in multiomic and metadata analysis using next generation tools.

The study includes 476 patients (302 patients with clinical samples, 170 episodes of sepsis, 25% with bacteremia) recruited in two periods: pre-COVID confinement (March 2019-February 2020) and June 2021-January 2022. The second phase of the study was implemented during the project to increase the sample size and to avoid the underestimation of sepsis with a respiratory portal of entry (high prevalence in the analysis of the first data) and to analyze the drastic change in the pattern of hospital infection and sepsis observed during the COVID pandemic. The activities carried out correspond to objectives #1 (Categorization of the intestinal and respiratory microbiota using taxonomic markers of group, species, gene), #2 (Identification of the diversity of bacterial species with high invasiveness capacity), #3 (Development of platforms for metadata analysis).

In objective #1, the metagenomic patterns (16SRNA, enterotypes, resistotypes) of the intestinal and respiratory microbiome of all septic patients was characterized, as well as the "poolings" of patients admitted each week. The results revealed dysbiosis patterns associated with several colonization pressure indices and, eventually, the endogenous acquisition of microorganisms. In objective #2, we detect a predominance of enterobacteria in microbiotas of septic and non-septic patients and also in patients with certain nosocomial infections (with a predominance of Escherichia coli in newly admitted patients and of Klebsiella and other enterobacteria, Serratia, Pseudomonas and Chryseobacteria, in long-term hospitalized patients) and events of bystander selection. Subpopulations were associated with the infection episodes, and the transmission routes of these pathogens were linked to both endogenous and exogenous acquisition from biotic and abiotic reservoirs. Within objective #3, we generated a metadata collection platform (clinical, microbiological, pharmacy, protocol and intervention data including electronic medical records) was created using the REDCap software, and a metadata analysis platform with ADONIS, SparCC, and GLMM-MiRKA software to explain the impact of metadata variables on differences in taxonomic composition between specific groups. Objective #4 aims to create a predictive model of infection and depends on the achievement of the first three, currently in progress.